Description of problem:
======================
stale glusterfs processes are getting created when we do a glusterd restart while quota is being enabled and disabled on a volume in loop.


Version-Release number of selected component (if applicable):
---------------
3.8.4-54.14

How reproducible:
---------------
always

Steps to Reproduce:
1.have a 1x3 volume started
2.open a terminal say t1 for n1, and do glusterd restart in loop, with 10sec gap between iterations
(#while true;do service glusterd restart;sleep 10;done)
3.from terminal t2 login to n1 and do quota enable/disable in loop
(#while true; do gluster v quota ntap enable;sleep 0.5;gluster v quota ntap disable --mode=script; done)


Actual results:
---------------
you will see many stale glusterfs procs only on n1

Expected results:
----------------
should not be seeing stale processes


Additional info: